PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度
Commits on main
align auto_quantizer with main branch in Transformers (#437)
7d27f67 WeiweiZhang1 committed 1 year ago
Fix packing hang, torch compile and force to fp16 at exporting (#430)
40ab982 Wenhua Cheng committed 1 year ago
speedup packing stage for autogptq and autoawq format (#407)
937d019 Wenhua Cheng committed 1 year ago